her. This structural clarity is vital when variables contain complex data, ensuring Claude correctly attributes the metadata (like the source) to the correct piece of content.","masteryContext":""},{"key":"D","isCorrect":false,"text":"XML tags prevent the model from using outside knowledge.","explanation":"XML tags do not inherently block the model's internal knowledge. While they help the model focus on the provided context, preventing the use of outside knowledge requires specific negative constraints and grounding instructions in the system prompt, rather than just the presence of XML delimiters.","masteryContext":""}],"correctOptionKey":"C","correctOptions":["C"],"correctAnswerText":"","isMultiSelect":false,"mainExplanation":"XML tags provide a clear, unambiguous structure that helps Claude identify where one variable ends and another begins. This is especially important when the variables contain unstructured text that might include characters like colons or dashes, which could confuse the model if simple text labels were used as delimiters.","examTrap":"Candidates often incorrectly assume XML tags are for tokenization efficiency or model speed, rather than recognizing their primary purpose as clear, unambiguous delimiters for variable content within prompts.","examTip":null,"technicalDeepDive":"Claude models are trained on large amounts of XML-like markup, so tags such as `...` or `...` act as strong structural delimiters that help the model distinguish injected variable content from the surrounding instructions. This reduces prompt-injection risk and ambiguity when a variable contains text resembling instructions. Tags do not compress tokens, are not required by the Messages API, and do not disable outside knowledge; they are purely a prompting convention. A realistic scenario: a summarization template injects a user-supplied document; wrapping it in `` tags makes it far less likely the model treats sentences inside the document as new instructions, which matters for any application exposing prompt templates to untrusted input.","keyConcepts":[],"faqs":[],"eliminationLogic":null,"exhibitText":"\n\n{{SOURCE_1}}\n{{CONTENT_1}}\n\n","exhibitUrl":null,"hasExhibit":false,"isExhibitQuestion":false,"questionPairs":null,"hasMatchingPairs":false,"isMatchingQuestion":false,"relatedQuestions":[],"breadcrumbItems":null,"nextQuestionUrl":null,"previousQuestionUrl":null,"domainPracticeUrl":null,"certificationUrl":null,"domainUrl":null,"hasAnswerOptions":true,"updatedAt":"2026-09-17T09:57:40.617493+00:00","questionType":"multiple_choice"},{"id":"e148b6ad-9a63-4447-93dd-9d46a00dbdcc","slug":"a-financial-services-company-is-building-a-customer-support-e3bp4","certificationSlug":"anthropic-claude-architect-foundations","examCode":"CCAR-F","examName":"Claude Certified Architect - Foundations","questionText":"A financial services company is building a customer support assistant that needs to handle loan applications, account balance checks, and general FAQ queries. Which orchestration pattern should be used to minimize token usage while ensuring that the model uses specific system instructions for each distinct task?","difficulty":"medium","domain":"Agentic Architecture and Orchestration","domainSlug":"agentic-architecture-orchestration","topic":"","questionNumber":0,"totalQuestions":0,"options":[{"key":"A","isCorrect":false,"text":"The Parallelized Workflow pattern","explanation":"This pattern is intended for tasks that can be broken down into independent sub-tasks executed simultaneously. Since the user's query only requires one of the three service types at a time, running multiple parallel processes would lead to unnecessary computational waste and increased latency without providing any functional benefit for categorization.","masteryContext":""},{"key":"B","isCorrect":true,"text":"The Router pattern","explanation":"Routing allows the system to identify the specific domain of a query